Diseño e implementación de una aplicación web con machine learning para la detección temprana de tumores cerebrales mediante resonancias magnéticas

Abstract

This research project presents the development of a web application for the early detection of brain tumors using artificial intelligence and computer vision techniques. The system is based on a convolutional neural network (CNN) model trained with magnetic resonance imaging (MRI) scans, applying the VGG16 architecture to classify images into four categories: glioma, meningioma, pituitary tumor, and no tumor. The technical implementation includes a backend developed with Flask in Python, a frontend designed with HTML, CSS, and Bootstrap, and an analysis module built in Jupyter Notebook. In addition to providing a preliminary diagnosis, the interface incorporates educational resources, such as an interactive 3D brain model and personalized recommendations based on the output. The results demonstrate satisfactory model performance, highlighting its potential as a support tool in scenarios with limited access to specialists and advanced diagnostic equipment.